David Sacks claims there's 'substantial evidence' that DeepSeek used OpenAI's models to train its own | TechCrunch
LOL
https://techcrunch.com/2025/01/28/david-sacks-claims-theres-substantial-evidence-that-deepseek-used-openais-models-to-train-its-own/Open linkView original on lemmy.ml124
Comments62
So the same people who have no problem about using other people's copyrighted work, are now crying when the Chinese do the same to them? Find me a nano-scale violin so I can play a really sad song.
That's obviously a cello.
Maybe a cello if it was human grade. But that’s tardigrade.
Pedant time: That's microscale not nanoscale.
You can shoot me now, it's deserved.
Can you put a liuqin in there?
Planck could not scale small enough.
You more elegantly said what I came to say.
Stealing from thieves is not theft
Yes it is. Although I personally have far less moral objections to it.
To elaborate:
OpenAI scraped data without permission, and then makes money from it.
Deepseek then used that data (even paid openai for it), trained a model on that data, and then releases that model for anyone to use.
While it's still making use of "stolen data" (that's a whole semantics discussion I won't get into right now), I find it far more noble than the former.
So it's more like Deepseek bought stolen items from the shady OpenAI pawnshop.
And then handed it out for free, yes. Although I'm sure they're making money out of it as well - obviously they're not saints.
Came to say something similar. Like I give a fuck that OpenAI's model/tech/whatever was "stolen" by Deepseek. Fuck that piece of shit Sam Altman.
"Recieving stolen goods" is prosecutable.
It's a lesser crime than the original theft though.
Cry me a fucking river, David.
Oh are we supposed to care about substantial evidence of theft now? Because there’s a few artists, writers, and other creatives that would like to have a word with you…
Bruh, these guys trained their own AI on so called "puplicly available" content. Except it was, and still is, completely without consent from, or compensation to said artists/bloggers/creators etc.. Don't throw rocks when you live in a glass house 🤌
Another reason why I use Andi, because it don¡t gut copyright content to the own knowledge base, it's is a search assistant, not a chatbot like others, it search by the concept and give a direct answer to your question, listing also the links of the sources and pages where it found the answers. It's LLM is only made to "understand" (to call it something) your question to search pages que contain information about it and to understand the content to be capable to summarize it. There isn't third party or copyright content in the LLM. It's knowledge is real time web content like any other search engine. Even in it's (reduced) chat capabilities, always show the sources where it found it's answers.
Traditional search works with keywords, listing thousends of pages where appears this keyword, that means that 99% of the list has nothing to do with what you are looking for, this is the reason why AI searches give a better result, but not Chatbots, which search the answers in a own knowledge base and invent answers if not.
Womp womp. I'm sure openAI asked for permission from the creators for all its training data, right? Thief complains about someone else stealing their stolen goods, more at 11.
OpenAI's mission statement is also in their name. The fact that they have a proprietary product that is not open source is criminal and should be sued out of existence. They are now just like the Sun Micro after Apache was made open sourced; irrelevant they just haven't gotten the memo yet. No company can compete against the whole world.
i agree FOSS is the way to go, and that OpenAI has a lot to answer for… but FOSS is not the only way to interpret “open”
the “open” was never intended as open source - it was open access. the idea was that anyone should have access to build things using AI; that it shouldn’t be for only megacorps who had the pockets to train… which they have, and still are doing
they also originally intended that all their research and patents would be open, which i believe they’re still doing
Your point OpenAI? Weren't you part of the group saying training AI wasn't copyright infringement? Not so happy when it's your shit being copied? Huh. Weird.
The only concern is how much the cost of training the model changes if it got a significant kickstart from previous, very-expensive training. I was interested because it was said to be comparable for a fraction of the cost. "Open"AI can suck sand.
When you can't win, accuse them of cheating.
But, but they committed the copyright infringement first. It's theirs. That's totally unfair. What are tech bros going to do? Admit they are grossly over valued? They've already spent the billions.
so he's just admitting that deepseek did a better job than openai but for a fraction of the price? it only gets better.
It's funny that they did all that and open-sourced it too. Like some kid accusing another to copy their homeworks while the other kid did significantly better and also offered to share.
Me pretending to care about David Sacks claim:
Here's the thing... It was a bubble because you can't wall off the entire concept of AI. This revelation was just an acceleration displaying what should've been obvious.
There are many many open models available for people to fuck around with. I have in a homelab setting, just to keep abreast of what is going on, get a general idea how it works and what its capable of.
What most normie followers of AI don't seem to understand is, whether you're doing LLM or machine learning object detection or something, you can get open software that is "good enough" and run it locally. If you have a raspberry pi you can run some of this stuff, and it will be slow, but acceptable for many use cases.
So the concept that only OpenAI would ever hold the keys and should therefore have massive valuation in perpetuity, that is just laughable. This Chinese company just highlighted that you can bruteforce train more optimized models on garbage-tier hardware.
Yes, AI arrived to stay, that is fact, also soon there will be a unified global AI (see Stargat Project) which will make obsolet all other LLM. A $500 Billon project. The only thing we can do is to use it in the most intelligent way, avoiding it will be impossible.
I couldn't give less of a fuck.
Open AI stole all of our data to train their model. If this is true, no sympathy.
That is what I mean, it's a difference between an AI with robbed content in its knowledge/lenguage base and an AI assistant which only search iformation in the web to answer, linking to the corresponding pages. Way more intelligent and ethic use of an AI.
FUD, just to distract from the crushing multibillion dollar defeat they’ve just been dealt. First stage of grief: denial. Second: anger. Third: bargaining. We’re somewhere between 2 and 3 right now.
Nope, it's definitely true, but sensationalism. Almost all models are trained using gpt
Copycat gets copycatted.
Yeah, and? WTF are you chuckle fucks gonna do about it? Whine and complain without a hint of irony? Because that's all this is. Because there's not a goddamn thing you can do about it, and you hate that. Not to mention the hit to your nearly bottomless wallets. Cry more emo kid, your suffering sustains me.
It's only ok when we do it, cause we're the good guys!
I think chatgpt is more self aware than OpenAI is.
Yes, so what?
https://stratechery.com/2025/deepseek-faq/
Who the fuck cares? They're all doing this.
They're obviously trying so hard for regulatory capture in the states it's embarrassing.
Well you can't run openai's models yourself so pretty sure deepseek would've had to pay for API access. How is that stealing again?
They’re eating each other.
Exactly. This is the end. The companies eat each other while we suffer.
Whoop de doo
So what? It's absolutely true and makes absolutely no difference to anyone
I still prefer Andisearch over all others, it was rhe first AI search long before all other were released, with own LLM not a copy from others.
It really doesn't matter which AI you prefer to use. You're wrong for using AI period.
you're wrong for using writing. Writing leads to laziness and forgetfulness. Future generations will hear much without being properly taught and will appear wise but not be so.
This is ALSO a garbage tier take. You have assumed your opponents opinion is based on AI making people dumb. You're providing them with ammo. There are many possibilities for why they think this. As far as you know, your statement is entirely irrelevant to the person you're responding too.
not my words, i was referring to Plato's Phaedrus to make fun of the "ai bad" comment. I'm sorry. Forgive me.
Sick! Context is king and I can pickupwhatyoure putting down here.
This is a garbage tier take. You use spell correction. Thats a form of computerized decision making, which is artificial intelligence. You MAY have a point if you're referring to LLMs but thats incredibly arguable, and you hasn't stated any reasoning behind your opinion.
Bullshit. You haven't made any argument at all. You just supported a position without any backing behind it. So your take is garbage. I don't use spell correction because I don't need it, as I understand the language.
Use of ai is not immoral in and of itself. That argument is made in my response. I supported that argument by explaining why AI isn't immoral, in that its a useful tool.
"I don't use spell correction because I don't need it, as I understand the language."
Run-on sentence. You are using a comma as a seperator between two fully separate thoughts. Thats the wrong punctuation to accomplish the task you want.you should either use a semi-colon or a conjunction, or just make them separate sentences.
Your understanding of the language is clearly perfect and any tool that provides feedback when you make mistakes is clearly unnecessary.
Ok, so basically you're saying you're a grammar Nazi? Why? You still haven't made any argument for your position. You're just a dick. What a waste of time and data. Congrats for that, I guess?
My argument, which has ready been made twice, is computer decision making is not immoral. Thats ridiculous. Its incredibly useful in thousands of ways. All of this is AI.
You ACTUALLY make no argument. "AI is bad. If you use ai, you're evil."
That is not an argument, just an opinion, and has no support.
I'm using spellcheck as an example because its relevant.
If you continue to believe I've made no argument, well, I guess your reading compression isnt up to where it would need to be to understand basic human interaction.
You are of course welcome to have and express your opinion, but when its just a shit take and its unsupported dont get pissed about getting called out.
Again you fail to make any point or support any argument. Congratulations, you made words. But you haven't even made it clear what side you're on. You're like a fucking late night ad show. You're the human version of QVC. You haven't even made it clear what your problem is with me, let alone my initial point. You can eat my ass or stfu. Up to you.
Good luck suing them